Widely used general anesthetics require administration by highly trained and vigilant experts, in part because they produce many undesirable effects and can be lethal at clinically used doses, particularly in elderly patients an those with systemic diseases. The safest (highest animal LD50/ED50 ratios) anesthetics, etomidate and alphaxalone, are both potent and selective modulators of GABAA receptors that also display stereoselectivity. These and similar drugs represent optimal starting points for developing improved and safer anesthetics. The overall aim of this renewal proposal is to define the interactions of etomidate, alphaxalone, propofol (a potent but less safe anesthetic) and a novel potent barbiturate (mTFD-MPAB) at their sites on pentameric (?1)2(3)2?2L GABAA receptors, and to clarify conformational changes in these sites that couple anesthetic binding to functional effects. Our working hypothesis is that these drugs act at distinct sites within the fiv inter-subunit transmembrane clefts formed by M1, M2, and M3 domains, and that these clefts expand during receptor activation and desensitization. Our studies to date have identified novel contact points for etomidate, propofol, alphaxalone, and mTFD-MPAB within multiple inter-subunit clefts. We also have engineered cysteines into these sites to reveal GABA-enhanced sulfhydryl modification and GABA-dependent anesthetic protection. We have also developed quantitative allosteric models that explain multiple GABAAR mutant functional phenotypes, including subsite contributions to each drug's actions. In Aim 1, we will identify which of the fiv inter-subunit sites interacts with each anesthetic, and which subsites transduce drug binding into enhanced receptor gating. We will use cysteine-substitutions in transmembrane domains M1, M2, and M3 of each subunit, along with chemical modification, protection, and electrophysiology, to define the binding clefts and residues closest to each anesthetic. Aim 2 will test the hypothesis that GABA activation enlarges all five inter-subunit transmembrane clefts in concert. We will examine anesthetic subsite interactions using cysteine accessibility assays and estimate the effective cut-off size of the anesthetic binding pockets using variable size mutations and cysteine adducts for three anesthetics that bind in the same inter-subunit clefts. We will also use novel sulfhydryl-modifying anesthetic derivatives and cysteine substituted residues to further define the binding orientation of stereoselective anesthetics. Aim 3 will test whether anesthetics influence GABAA receptor rates of desensitization and whether the shape of the inter-subunit clefts differs between open and desensitized states, again based on cysteine accessibility. These studies will use submillisecond solution switching and patch-clamp electrophysiology. Data will be analyzed using functional allosteric models and further interpreted in the context of evolving GABAAR structural models.